THAT4301 compressor

[iprovlek]

Need help with gain reduction with this THAT 4301 compressor!

I bought from DIY-TUBES.com!

Can be 1mA meter,or another meter use here in this project!?

If meter can be use what I must connect to work for gain reduction!?

Thanks

 

[hazel]

Need help with gain reduction with this THAT 4301 compressor!

I bought from DIY-TUBES.com!

Can be 1mA meter,or another meter use here in this project!?

If meter can be use what I must connect to work for gain reduction!?

Thanks

Looking at schematics http://diy-tubes.com/image/data/manuals/that4301-v1.5.2.pdf you can see first page bottom right corner, after Ratio wiper pot there's R19 15K optional resistor, just replace it with your 1mA meter and it should work

 

[abbey road d enfer]

Need help with gain reduction with this THAT 4301 compressor!

I bought from DIY-TUBES.com!

Can be 1mA meter,or another meter use here in this project!?

If meter can be use what I must connect to work for gain reduction!?

Thanks

Looking at schematics http://diy-tubes.com/image/data/manuals/that4301-v1.5.2.pdf you can see first page bottom right corner, after Ratio wiper pot there's R19 15K optional resistor, just replace it with your 1mA meter and it should work

No, it wouldn't work. The load presented by the meter would be utterly wrong for the circuit. The voltage here is -6mV per dB of Gain Reduction. It is necessary to insert a buffer there; in fact, more precisely, it needs be a voltage-to-current converter.
One has to decide first what scale is targetted. Generally, people opt for about 30-40dB GR for full-scale deviation.
Let's say 30dB with a 1mA meter.
You need a converter with a characteristic of 1mA for 180mV (6x30). It can be as simple as a simple 180 ohm resistor.
This resistance must include the meter's internal resistance, which is often higher than that.
So there should be some gain, which the buffer would also provide.
It's a problem with more unknows than equations, so there is an infinity of solutions.
Let's see:
r is the meter's internal resistance in ohms
R is the series resistance in ohms
I is the full-scale current in A
GR is the maximum GR corresponding to full-scale deviation in dB
G is the buffer's gain (no dimension, just a number)

We get:
I=6. 10e-3.G.GR/ (R+r) or
R=[6. 10e-3.G.GR/I] - r
It's a problem with more unknows than equations, so there is an infinity of solutions.
For the buffer, you may use any old opamp (TL0, 741...).
You must choose a value for R+r that doesn't imply too high output voltage from the opamp. For a 1mA full-scale meter, that would be about 10k.

 

[darkus]

hazel

Did you ever finish this build?

I just purchased two pcbs and gonna start working on this soon. Im hoping this simple comp will give me that dbx160xt sound. Ive listened to your clips and i hear this compressor is distorting pretty fast?

Hi Darkus!

Compressor is finished and working but i'm designing front panel.

Samples in my Soundcloud are an example of Time Constant Capacitor values at THAT4301 pin 5.

10uF cap for RMS behavior and change them to taste for some distorted sound, i personally use crushed sound with blend control for parallel compression and RMS with SOFT KNEE for insert use.

This compressor doesn't only distorts, this compressor doesn't only sounds smooth and transparent...it does both things!

You can just build it using stock values or mod it to your taste, there's plenty of info in this thread (HPF SC, manual Attack and Release, HARD/SOFT knee, etc)

Regards,

Joan

Do you have any clean compression soundclips? Im looking to make this build to sound like DBX160 series...pop and smack for kicks and snares.

-Markus

 

[hazel]

No, it wouldn't work. The load presented by the meter would be utterly wrong for the circuit. The voltage here is -6mV per dB of Gain Reduction. It is necessary to insert a buffer there; in fact, more precisely, it needs be a voltage-to-current converter.
One has to decide first what scale is targetted. Generally, people opt for about 30-40dB GR for full-scale deviation.
Let's say 30dB with a 1mA meter.
You need a converter with a characteristic of 1mA for 180mV (6x30). It can be as simple as a simple 180 ohm resistor.
This resistance must include the meter's internal resistance, which is often higher than that.
So there should be some gain, which the buffer would also provide.
It's a problem with more unknows than equations, so there is an infinity of solutions.
Let's see:
r is the meter's internal resistance in ohms
R is the series resistance in ohms
I is the full-scale current in A
GR is the maximum GR corresponding to full-scale deviation in dB
G is the buffer's gain (no dimension, just a number)

We get:
I=6. 10e-3.G.GR/ (R+r) or
R=[6. 10e-3.G.GR/I] - r
It's a problem with more unknows than equations, so there is an infinity of solutions.
For the buffer, you may use any old opamp (TL0, 741...).
You must choose a value for R+r that doesn't imply too high output voltage from the opamp. For a 1mA full-scale meter, that would be about 10k.

Thanks for the info, GR meter works in the circuit but didn't know all this math

Do you have any clean compression soundclips? Im looking to make this build to sound like DBX160 series...pop and smack for kicks and snares.

-Markus

Hi Markus,

I don't know if it's possible to make a 4301 to sound like a 160, i'd call it "the invisible compressor".

I recently moved to a new apartment and i'm just re-building my studio,

I'll try to post some clips this weekend

Regards,

Joan

 

[iprovlek]

hazel does your meter works on your 4301compressor!?

You use DIY-TUBE kits that 4301 compressor?

 

[hazel]

hazel does your meter works on your 4301compressor!?

You use DIY-TUBE kits that 4301 compressor?

I used a Demo Board from THAT Corp but it's the same, compare both schematics. what's the issue with your 1mA meter?? don't work at all? needle gets stucked all way down or up with light or no compression?

 

[iprovlek]

Thanks for response hazel!

Needle gets stucked all way down with no compression,if I solder meter to PCB!

Even I try DIY-TUBE kits VU meter buffer,and nothing!


I dont know what do to!

 

[abbey road d enfer]

Needle gets stucked all way down with no compression,if I solder meter to PCB!

That means there is something seriously wrong about the operation of the circuit, or maybe you have connected the meter incorrectly.
With the ratio pot at minimum (1:1), there should be no voltage at the meter. Can you check the voltage at R20 and R24?

 

[iprovlek]

Thanks Abbey on response!

Yesterday I try this meter
http://export.farnell.com/multicomp/mc34h-0-100/meter-edge-indicating-100ua/dp/143509

And I solder meter backwards and he started work!
But no properly, It show light compression,even it barely move it!
strange!

I dont know is there option to get this meter more sensitive!?
I will check R24 and R20!

 

[abbey road d enfer]

Thanks Abbey on response!

Yesterday I try this meter
http://export.farnell.com/multicomp/mc34h-0-100/meter-edge-indicating-100ua/dp/143509

And I solder meter backwards and he started work!
But no properly, It show light compression,even it barely move it!
strange!

I dont know is there option to get this meter more sensitive!?
I will check R24 and R20!

can you measure the resistance of the meter? As I wrote earlier, the voltage here is very low, about 120mV for 20dB GR; if the internal resistance of the meter is too high, this low voltage will not produce enough current to drive the meter.

 

[iprovlek]

You right Abbey!
With 1mA I get nothing,but with 100uA meter get some moving!

I think that I need use more sensitive meter!

I try to measure R20 and R24 but dont have lucky!

Internal resistance of 100uA is 2ohms but with 1mA is 200ohms!

 

[abbey road d enfer]

You right Abbey!
With 1mA I get nothing,but with 100uA meter get some moving!

I think that I need use more sensitive meter!

I don't believe you can find one easily.

I try to measure R20 and R24 but dont have lucky!

Internal resistance of 100uA is 2ohms but with 1mA is 200ohms!

That is very strange. Couldn't it be that the 100uA is 2 kohm?
The 1mA, with 200 ohm internal resistance needs 200mV for full scale, which corresponds to 33dB GR, but it would take 25V at the output of OAG$1, which is impossible.
Now, if the 100uA meter has an internal resistance of 2kohm, it still wants to see 200mV for full-scale, but it takes only about 3 V at the output of OAG$1, which can be achieved, but makes the ratio completely wrong, by a factor 3 or 4.
As I wrote earlier, you can't just connect the meter directly, you must use a buffer.

 

[iprovlek]

Thanks Abbey!

100uA have 2ohms not 2Kohms of internal resistance!

So I need buffer for this!

I bought this VU-meter buffer
http://diy-tubes.com/vu-buffer-pcb.html

Can be this buffer good for this purpose!?

 

[abbey road d enfer]

Thanks Abbey!

100uA have 2ohms not 2Kohms of internal resistance!

Are you 100% positive? That seems unlikely.

So I need buffer for this!

Yes.

I bought this VU-meter buffer
http://diy-tubes.com/vu-buffer-pcb.html

Can be this buffer good for this purpose!?

You could make it work, but it would take a lot of modifications. You need a simple DC buffer powered from dual rails (+ and -).

 

[iprovlek]

Thanks Abbe!
Yes Im sure!
I measured twice!

Is this schematics could be find on internet!
Ill need value of components!?

Thanks a lot Abbe

 

[iprovlek]

I find this!

 

[iprovlek]

And this!
True buffer "follower"!

 

[abbey road d enfer]

I find this!

This one is good because by varying the resistors its gain can be adjusted.

 

[abbey road d enfer]

And this!
True buffer "follower"!

That one is NOT good because the voltage gain may be insufficient.